Real time analysis with development, simulation and validation of discrete element method models for tumbling mill charge motion and liner wear

Tumbling mills, which consume around 6% of world's electric power, though inefficient (use 1% of the supplied energy) have been used in mineral processing industry for over a century. Harsh internal mill environments have prevented the access to in-mill dynamics. This situation has led researchers to develop models based on the Discrete Element Method (DEM) to predict internal mill dynamics. But before DEM models can be used in the mineral processing industry, their rigorous validation is essential.; In this thesis, we have carried out real time analysis of the DEM model to enable online availability of the mill parameters for the mill operators. We have completed rigorous validation of a DEM model based on nine mill parameters by comparing the predicted parameters under varying operating conditions of a 30" pilot mill and some simulated industrial mills. Further, we have also developed a simulator to predict the wear on the tumbling mill liner profile.; Results show that the charge motion simulator used is faster than real time. The validation exercise showed that improvements to charge motion are possible, liner wear prediction is feasible but affects real-time simulator behaviour. Based on these results, it is recommended to continue the improvement of the charge motion model as well as the liner wear model.